TMEM106C contributes to the malignant characteristics and poor prognosis of hepatocellular carcinoma

Abstract

Transmembrane protein (TMEM) is a kind of integral membrane protein that spans biological membranes. The functions of most members of the TMEM family are unknown. Here, we conducted bioinformatic analysis and biological validation to investigate the role of TMEM106C in HCC. First, GEPIA and Oncomine^TM were used to analyze TMEM106C expression, which was verified by real-time PCR and western blot analyses. Then, the biological functions of TMEM106C were explored by CCK8 and transwell assays. The prognostic value of TMEM106C was analyzed by UALCAN. LinkedOmics was used to analyze TMEM106C pathways generated by Gene Ontology. A protein-protein interaction network (PPI) was constructed by GeneMANIA. We demonstrated that TMEM106C was overexpressed in HCC and that inhibition of TMEM106C significantly suppressed the proliferation and metastasis of HCC through targeting CENPM and DLC-1. Upregulation of TMEM106C was closely correlated with sex, tumor stage, tumor grade and prognosis. Overexpression of TMEM106C was linked to functional networks involving organelle fission and cell cycle signaling pathways through the regulation of CDK kinases, E2F1 transcription factors and miRNAs. Our data demonstrated that TMEM106C contributes to malignant characteristics and poor prognosis in HCC, which may serve as a prognostic biomarker and potential therapeutic target.

Keywords: TMEM106C; bioinformatics; hepatocellular carcinoma; metastasis; proliferation.

Figure 1

TMEM106C transcription in tumor tissues HCC (GEPIA and Oncomine). (A) Boxplot showing TMEM106C mRNA levels in TCGA and GTEx from GEPIA. *P < 0.05. (B–F) Boxplot showing TMEM106C mRNA levels in Chen Liver, MasLiver, RoesslerLiver, RoesslerLiver2, and Wurmbach Liver datasets from Oncomine. (T=tumor, N=normal).

Figure 2

TMEM106C transcription in subgroups of HCC patients, stratified by gender, age, race, tumor stage and tumor grade (UALCAN). (A) Relative level of TMEM106C in normal liver and LIHC samples. (B) Boxplot showing the relative expression of TMEM106C in healthy controls and male or female LIHC patients. (C) Relative level of TMEM106C in healthy controls of any age and LIHC patients of different age periods. (D) Relative level of TMEM106C in healthy controls of any ethnicity and LIHC patients of Caucasian, African-American or Asian ethnicity. (E) Boxplot showing the relative expression of TEM106C in healthy controls and LIHC patients in different stages. (F) Relative level of TMEM106C in healthy controls and LIHC patients with grades 1, 2, 3 or 4. 31 Data are shown as the mean ± SE.

Figure 3

Expression validation, functional exploration and prognostic value of TMEM106C in HCC. (A) Relative expression level of TMEM106C in 10 pairs of HCC samples (tumor tissues and adjacent normal liver tissues), as assessed by real-time PCR. (B) The protein expression level of TMEM106C in 10 pairs of HCC samples, as assessed by western blot. (C) TMEM106C IHC staining in HCC and adjacent normal liver tissue. 400×. (D) The protein expression level of TMEM106C in the normal liver cell line L02 and in different HCC cell lines. The cell line SMMC-7721 was selected for further study. (E) siRNAs targeting TMEM106C were transfected into SMMC-7721 cells for 48 h, and then all cell lysates were harvested for western blotting. (F) TMEM106C levels under si-TMEM106C, pcDNA-TMEM106C plasmid or si-TMEM106C plus pcDNA-TMEM106C plasmid assessed by western blot. (G) si-TMEM106C (50 nM) and pcDNA-TMEM106C plasmid plus si-TMEM106C were transfected into SMMC-7721 cells and were compared to untransfected control cells. Every 12 h, cell numbers were measured by CCK8 assay. NC represents pcDNA3.1, si-NC, and blank control, which were proven to not be different from each other. P < 0.01. 32 (H) Transwell migration and invasion assays of SMMC-7721 cells after transient transfection with si-TMEM106C or pcDNA-TMEM106C plasmid plus si-TMEM106C or not. The migration and invasion cell numbers are shown in histograms (mean ± SD). NC represents pcDNA3.1, si-NC, and blank control, which were proven to not be different from each other. 400×. *P < 0.05, **P < 0.01. (I) The overall survival rates of 364 HCC patients were compared between the TMEM106C high and low expression groups using Kaplan-Meier analysis (GEPIA). (J) The disease-free survival rates of 364 HCC patients were compared between the TMEM106C high and low expression groups using Kaplan-Meier analysis (GEPIA).

Figure 4

TMEM106C expression correlated genes in HCC (LinkedOmics and GEPIA). (A) Correlations between TMEM106C and differentially expressed genes in LIHC from LinkedOmics (Pearson test). Red indicates positively correlated genes, and green indicates negatively correlated genes. (B, C) Heat maps showing the top 50 genes positively and negatively correlated with TMEM106C in LIHC. (D–F) The scatter plots show the Pearson correlation of TMEM106C expression with the most positively correlated genes: CENPM, UBE2T and CDT1 (GEPIA). (G–I) The scatter plot shows the Pearson correlation of TMEM106C expression with the most 33 negatively correlated genes: ZC3H13, DLC1 and RANBP3L (GEPIA).

Figure 5

The verification of TMEM106C potentially related target genes in HCC. (A) Relative level of CENPM in normal liver and LIHC samples (UALCAN). (B) Relative level of DLC-1 in normal liver and LIHC samples (UALCAN). (C) Relative expression level of CENPM in 10 pairs of HCC samples (tumor tissues and adjacent normal liver tissues), as assessed by real-time PCR. (D) Relative expression level of DLC-1 in 10 pairs of HCC samples (tumor tissues and adjacent normal liver tissues), as assessed by real-time PCR. (E) si-CENPM (50 nM) and pcDNA-TMEM106C plasmid plus si-CENPM were transfected into SMMC-7721 cells. Every 12h, the cell number was measured by CCK8 assay. NC represents pcDNA3.1, si-NC, and blank control, which were proven to not be different from each other. 400×. P < 0.001. (F) pcDNA-TMEM06C and pcDNA-DLC-1 plus pcDNA-TMEM06C plasmid (or no plasmid control) were transfected into SMMC-7721 cells. Every 12 h, the cell number was measured using the CCK8 assay. NC represents pcDNA3.1 and the blank control, which was proven to not be different from each other. P < 0.001. (G) Transwell migration and invasion assays of SMMC-7721 cells after transient transfection with pcDNA-TMEM06C, si-CENPM, pcDNA-TMEM06C and si-CENPM, pcDNA-DLC-1 plus 34 pcDNA-TMEM06C plasmid (or no plasmid control). The migration and invasion cell numbers are shown in histograms (mean ± SD). NC represents pcDNA3.1, si-NC, and blank control, which were proven to not be different from each other. *P < 0.05, **P < 0.01.

Figure 6

Significantly enriched GO annotations and KEGG pathways of TMEM106C in HCC analyzed by GSEA (LinkedOmics). (A) Cellular components. (B) Biological processes. (C) Molecular functions. (D) KEGG pathway analysis. FDR=0. (E) KEGG pathway annotations of the cell cycle pathway.

Figure 7

Interaction network of associated genes generated by GENEMANIA. (A) The PPI network and functional analysis consisted of the TF-kinase-miRNA-TMME106C interaction. These genes were linked by different colors indicating the following relation: coexpression, physical interaction, colocalization, pathway or predicted, while the different colors for the network nodes indicate the biological functions of the set of enriched genes.